estimagic is a Python package that provides high-quality and user-friendly tools to fit large scale empirical models to data and make inferences about the estimated model parameters. It is especially suited to solve difficult constrained optimization problems.

estimagic provides several advantages over similar packages, including a unified interface that supports a large number of local and global optimization algorithms and the possibility of monitoring the optimization procedure via a beautiful interactive dashboard.

estimagic provides tools for nonlinear optimization, numerical differentiation and statistical inference.



  • estimagic wraps all algorithms from scipy.optimize and many more become available when installing optional dependencies.

  • estimagic can automatically implement many types of constraints via reparametrization, with any optmizer that supports simple box constraints.

  • estimagic encourages name-based parameters handling. Parameters are specified as pandas DataFrames that can have any kind of single or MultiIndex. This is especially useful when specifying constraints.

  • The complete history of parameters and function evaluations are saved in a database for maximum reproducibility and displayed in real time via an interactive dashboard.


Numerical differentiation

  • estimagic can calculate precise numerical derivatives using Richardson extrapolations.

  • Function evaluations needed for numerical derivatives can be done in parallel with pre-implemented or user provided batch evaluators.

Statistical inference

  • estimagic provides asymptotic standard errors for maximum likelihood and method of simulated moments.

  • estimagic also provides bootstrap confidence intervals and standard errors. Of course the bootstrap procedures are parallelized.